Dynamic gating scheduling method for time-sensitive network
By enabling switching nodes to autonomously perform incremental gating scheduling calculations and collaborative configuration distribution, the computational latency and network communication overhead issues of traditional TSN systems when dynamically adding new service flows are resolved, achieving efficient dynamic service expansion and deterministic transmission.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-03-27
AI Technical Summary
Traditional Time-Sensitive Networking (TSN) systems rely on a centralized network controller for global replanning when faced with dynamically added service flows. This leads to increased computational latency and network communication overhead, making it difficult to balance the real-time nature of new service flows with the transmission guarantee of existing service flows in multi-hop, multi-level switching node topologies. The system also suffers from insufficient scalability in dynamic scenarios.
Without relying on a centralized network controller for global replanning, the sending node generates new service flow request information, the switching node performs incremental gating scheduling calculations based on pre-stored static configuration data, and uploads the scheduling results level by level from bottom to top. The updated gating scheduling configuration is then distributed in a unicast manner to ensure that all switching nodes are consistent and effective.
It significantly reduces scheduling computational complexity, improves the response speed of new service flow access, enhances idle time slice utilization and scheduling success probability, avoids data conflicts caused by configuration asynchrony, and realizes dynamic service expansion.
Smart Images
Figure CN121750583A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of network bus, and particularly relates to a dynamic gating scheduling method of a time-sensitive network. BACKGROUND
[0002] With the rapid development of industrial automation, intelligent transportation, aerospace and unmanned systems, the real-time, reliability and determinacy requirements of network communication are increasingly improved. As a communication technology for industrial and critical task scenarios, Time-Sensitive Networking (TSN) can provide deterministic transmission guarantee on the basis of Ethernet, and realize low-latency, low-jitter and high-reliability transmission of business flow through time scheduling and gating mechanism.
[0003] The traditional TSN system usually relies on a centralized network controller (CNC) to perform global traffic planning and gating scheduling. In the network initialization stage, the CNC generates static configuration data offline according to the global network topology, link resource state and preset business flow demand, including the forwarding table of each switching node, the port gating list and the gating period information, and delivers them to the switching node to guide the forwarding and transmission of the business flow. Although this method can provide deterministic guarantee for known business flow, it has obvious shortcomings when facing dynamic new business flow.
[0004] Static configuration is only applicable to known business flow in the initialization stage, and cannot respond to new business demand generated in the running process of an aircraft, a drone or an industrial automation system in real time. The scheduling of new business flow usually relies on the global planning of the CNC, which has the problems of calculation delay and response lag. The centralized controller needs to perform global calculation and delivery, and with the increase of network size and the number of business flow, the calculation load of the CNC is significantly improved, and the delivery of global network configuration increases the network communication overhead, reduces the real-time response ability of the system, and in the multi-hop, multi-level switching node topology, the static gating configuration is difficult to balance the real-time of new business flow and the transmission guarantee of existing business flow, resulting in insufficient scalability of the system in dynamic scenarios. SUMMARY
[0005] The purpose of the application is to provide a dynamic gating scheduling method of a time-sensitive network, which can realize fast and deterministic scheduling of incremental business flow without relying on global re-planning of a centralized network controller.
[0006] The technical scheme adopted by the application is as follows:
[0007] A dynamic gating scheduling method of a time-sensitive network, comprising:
[0008] The sending node generates new service flow request information and sends the new service flow request information to the directly connected switching node;
[0009] The switching node, after receiving the new service flow request information, forwards the new service flow request information to the next hop switching node according to the network topology relationship, and stops forwarding the new service flow request information when the next hop of the new service flow is the switching node corresponding to the destination node.
[0010] Any switching node receiving the new service flow request information performs incremental gating scheduling calculation on the new service flow based on the static configuration data previously issued and stored by the centralized network controller without re-invoking the centralized network controller.
[0011] The switching node uploads the scheduling result to the upper switching node after completing the incremental gating scheduling calculation, and issues the updated gating scheduling configuration to the related switching nodes when the switching node located at the starting end of the path confirms that the new service flow is successfully scheduled on the corresponding whole forwarding path, so that each switching node performs data forwarding control according to the updated gating scheduling configuration. When any node returns a scheduling failure result, the sending node is fed back information that the new service flow is not schedulable.
[0012] In the case that the new service flow is successfully scheduled, the switching node directly connected with the sending node issues the corresponding service flow sending time configuration to the sending node, so that the sending node performs data sending according to the sending time configuration.
[0013] In a preferred scheme, the new service flow request information includes service flow type, source node identifier, destination node identifier, data frame length, frame number, and sending period parameters.
[0014] In a preferred scheme, the static configuration data is generated and issued offline by the centralized network controller. The centralized network controller, based on global network topology, link resource state, and preset service flow demand, uniformly plans the forwarding table items and gating list of each switching node in the network initialization stage, and issues the generated static gating scheduling configuration to each switching node, so that each switching node takes the static gating scheduling configuration as the constraint background when performing incremental gating scheduling calculation.
[0015] In a preferred scheme, the static configuration data includes network topology information, forwarding table information of each switching node port, gating list corresponding to each switching node port, and period information.
[0016] In a preferred scheme, the incremental gating scheduling calculation includes:
[0017] According to the network topology information, a forwarding path corresponding to the new service flow from the source node to the destination node is determined according to the shortest path principle, wherein the forwarding path is composed of a plurality of sequentially connected switching nodes;
[0018] According to the new service flow, a service flow order is generated according to a preset heuristic sorting rule;
[0019] In the gating list defined by the static configuration data, the idle time slice corresponding to each switching node port passed by the new service flow in the gating period is extracted, and the idle time slice is traversed according to the sorted service flow order to obtain a scheduling result.
[0020] In a preferred scheme, the new service flow is sorted and transmission time slots are sequentially assigned within the gating period defined by the static configuration data, wherein the sorting of the new service flow is based on the length of the forwarding path passed by the service flow, the data cache occupancy of the service flow, the maximum end-to-end delay allowed by the service flow, and the transmission period size of the service flow, so as to assign the idle time slice to the new service flow.
[0021] In a preferred scheme, the traversal of the idle time slice includes:
[0022] When the idle time slice of all relevant ports on the forwarding path can be assigned to the new service flow, an incremental gating scheduling configuration corresponding to the new service flow is generated;
[0023] When any switching node port cannot assign an idle time slice to the new service flow that satisfies the constraint condition, it is determined that the incremental scheduling of the new service flow fails.
[0024] In a preferred scheme, after completing the incremental gating scheduling calculation, each switching node uploads the scheduling result in a top-down manner, and when the switching node at the starting end of the forwarding path confirms that the new service flow is successfully scheduled at all switching nodes on the path, a cooperative gating scheduling configuration is triggered, and the updated gating scheduling configuration is unicast to the relevant switching nodes, and the configuration response mechanism is used to confirm that each switching node completes the configuration update.
[0025] In a preferred scheme, the transmission time configuration and the gating scheduling configuration are controlled in time sequence by a cooperative unicast method.
[0026] And a time-sensitive network dynamic gating scheduling terminal, comprising:
[0027] One or more processors;
[0028] A storage device having one or more programs stored thereon;
[0029] When one or more programs are executed by one or more processors, the one or more processors implement the time-sensitive network dynamic gating scheduling method.
[0030] The technical effects achieved by the present application are:
[0031] In the network initialization stage, the static configuration generation is completed by the centralized network controller, and in the running stage, the scheduling of the newly added service flow is independently completed by the switching node based on the existing static configuration, without the need to call the centralized controller again, thereby reducing the control link load, performing incremental gating scheduling calculation on the switching node side, planning only the newly added service flow, without affecting the gating configuration of the existing service flow, significantly reducing the scheduling calculation complexity, and improving the response speed when the newly added service flow is accessed.
[0032] The present application, through the heuristic sorting strategy based on path length, cache occupancy, end-to-end delay and sending cycle and other factors, makes the service flow with smaller resource occupation and more easily satisfied constraints participate in scheduling in priority, effectively improves the idle time slice utilization rate and the overall scheduling success probability, through the bottom-up scheduling result convergence mechanism and the cooperative configuration of the unicast mode, and combining the configuration response mechanism, realizes the consistent effect of the gating scheduling configuration between multiple switching nodes, avoids data conflict or timing exception caused by asynchronous configuration, and can complete dynamic service expansion without continuous dependence on the centralized controller, and is especially suitable for unmanned systems, avionics networks, industrial control networks and other application scenarios with high real-time and reliability requirements and relatively fixed network topology but dynamic service changes. BRIEF DESCRIPTION OF DRAWINGS
[0033] Figure 1 The present application provides a method flowchart. DETAILED DESCRIPTION
[0034] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.
[0035] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the concept of the present application, therefore the present application is not limited to the specific embodiments disclosed below.
[0036] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In a preferred embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.
[0037] Thirdly, the application is described in detail in combination with the schematic diagram, and in the detailed description of the embodiments of the application, the schematic diagram is only an example and should not limit the protection scope of the application herein.
[0038] Please refer to the accompanying Figure 1 As shown in the accompanying drawings, a time-sensitive network dynamic gating scheduling method is provided, comprising:
[0039] S1. A sending node generates new service flow request information and sends the new service flow request information to a directly connected switching node;
[0040] S2. After receiving the new service flow request information, the switching node forwards the new service flow request information to the next hop switching node according to the network topology relationship, and stops forwarding the new service flow request information when the next hop of the new service flow is the switching node corresponding to the destination node;
[0041] S3. Any switching node receiving the new service flow request information performs incremental gating scheduling calculation on the new service flow based on the static configuration data previously issued and stored by the centralized network controller without re-invoking the centralized network controller;
[0042] S4. After completing the incremental gating scheduling calculation, the switching node uploads the scheduling result to the upper switching node, and when the switching node located at the starting end of the path confirms that the new service flow is successfully scheduled on the corresponding entire forwarding path, the switching node issues the updated gating scheduling configuration to the related switching nodes, so that each switching node performs data forwarding control according to the updated gating scheduling configuration, and when any node returns a scheduling failure result, the sending node is fed back information that the new service flow is not schedulable;
[0043] S5. In the case that the new service flow is successfully scheduled, the switching node directly connected with the sending node issues the corresponding service flow sending time configuration to the sending node, so that the sending node performs data sending according to the sending time configuration.
[0044] It should be noted that the new service flow request information includes service flow type, source node identifier, destination node identifier, data frame length, frame number and sending period parameters.
[0045] Further, the sending time configuration and the gating scheduling configuration are controlled in time sequence by using a cooperative issuing mode.
[0046] As described in the above steps S1 to S5, when there is a new communication demand, the sending node generates a new service flow request information and sends it to the switching node directly connected thereto, the new service flow request information at least including a service flow type, a source node identifier, a destination node identifier, a data frame length, a frame number, and a sending period parameter, for describing the basic transmission constraint conditions of the new service flow. After receiving the new service flow request information, the switching node determines the next hop node of the new service flow according to the network topology relationship stored by itself, and forwards the new service flow request information to the next switching node along the topology path hop by hop. When it is determined that the next hop of the new service flow is the switching node corresponding to the destination node, the forwarding of the new service flow request information is stopped, so as to limit the processing range of the new service flow to only cover its actual forwarding path. After receiving the new service flow request information at any switching node, the switching node does not need to call the centralized network controller again, but performs incremental gating scheduling calculation on the new service flow based on the static configuration data generated offline by the centralized network controller and delivered in advance, the static configuration data including network topology information, forwarding table information of each switching node port, gating list and its period information, for serving as the constraint background of incremental scheduling. When performing the incremental gating scheduling calculation, the switching node first determines the forwarding path of the new service flow from the source node to the destination node according to the shortest path principle based on the network topology information, sorts the new service flow based on a preset heuristic sorting rule, the heuristic sorting rule taking the greedy algorithm as the core and comprehensively considering the length of the forwarding path passed through by the service flow, the data cache occupancy, the maximum end-to-end delay allowed, and the size of the sending period, so as to determine the priority order of the service flow participating in scheduling. The switching node extracts the idle time slices corresponding to each switching node port passed through by the new service flow in the gating period from the gating list defined in the static configuration data, and traverses and matches the idle time slices in the order of the sorted service flow. If the transmission time satisfying the transmission constraint conditions of the new service flow can be allocated in the idle time slices of all the related ports on the forwarding path, an incremental gating scheduling configuration corresponding to the new service flow is generated. If any switching node port cannot allocate an idle time slice satisfying the constraint conditions, it is determined that the incremental scheduling of the new service flow fails. After completing the local incremental gating scheduling calculation, each switching node uploads the scheduling result in a bottom-up manner, and when the switching node located at the starting end of the forwarding path confirms that all the switching nodes on the entire forwarding path have successfully scheduled the new service flow, a cooperative delivery process of the gating scheduling configuration is triggered, the updated gating scheduling configuration is delivered to the related switching nodes through unicast, and the completion of configuration update by each switching node is confirmed through a configuration response mechanism, so as to ensure the consistent effect of the gating configuration in time. In the case that the scheduling of the new service flow is successful, the switching node directly connected to the sending node further delivers the corresponding service flow sending time configuration to the sending node.The sending nodes transmit data according to the transmission time configuration. The transmission time configuration and gating scheduling configuration are distributed collaboratively for timing control to avoid service conflicts or timing mismatches caused by asynchronous configuration activation. During network initialization, the centralized network controller generates static configurations. During operation, the scheduling of new service flows is independently completed by the switching nodes based on existing static configurations, without needing to call the centralized controller again, thus reducing the load on the control links. By performing incremental gating scheduling calculations on the switching node side, only new service flows are planned without affecting the gating configurations of existing service flows, significantly reducing scheduling computation complexity and improving the response speed when new service flows are accessed. This is achieved through path length-based scheduling. A heuristic sorting strategy based on factors such as cache usage, end-to-end latency, and transmission cycle prioritizes service flows with lower resource consumption and easier-to-meet constraints for scheduling, effectively improving idle time slice utilization and overall scheduling success probability. Through a bottom-up scheduling result aggregation mechanism and unicast-based collaborative configuration distribution, combined with a configuration response mechanism, consistent gating scheduling configurations across multiple switching nodes are achieved, avoiding data conflicts or timing anomalies caused by configuration asynchrony. Dynamic service expansion can be completed without continuous reliance on a centralized controller, making it particularly suitable for application scenarios with high real-time and reliability requirements, relatively fixed network topologies, but dynamically changing services, such as unmanned systems, avionics networks, and industrial control networks.
[0047] In a preferred embodiment, static configuration data is generated and distributed offline by a centralized network controller. During the network initialization phase, the centralized network controller plans the forwarding table entries and gating lists of each switching node in a unified manner based on the global network topology, link resource status, and preset service flow requirements. The generated static gating scheduling configuration is then distributed to each switching node, so that each switching node uses the static gating scheduling configuration as a constraint background when performing incremental gating scheduling calculations.
[0048] It is worth mentioning that the static configuration data includes network topology information, forwarding table information for each switching node port, gating list and periodic information for each switching node port.
[0049] As described above, during the network deployment or initialization phase, the centralized network controller performs unified analysis and planning on the entire time-sensitive network based on the global network topology, link resource status, and preset service flow requirements. The global network topology describes the connection relationships and port connectivity between each switching node. The link resource status characterizes the bandwidth capacity and available time slot distribution of each link. The preset service flow requirements describe the types of services and their communication parameters that need to be supported during the initial operation phase. The centralized network controller performs unified offline planning processing on each switching node, generating corresponding forwarding table entries and gating lists. The forwarding table entries limit the forwarding path of data frames in the network, and the gating lists describe the opening and closing sequence of each switching node's ports within the gating period, thus forming a complete static gating scheduling configuration. The generated static gating scheduling configuration is controlled by the centralized network controller. The information is distributed to each switching node for local storage, ensuring that each switching node possesses network topology information, forwarding table information, and gating list and its periodicity information corresponding to its own port. During network operation, when new service flow demands arise, each switching node, when performing incremental gating scheduling calculations, no longer requests global scheduling from the centralized network controller. Instead, it directly performs calculations using the static gating scheduling configuration as the constraint background. When performing incremental scheduling, the switching node only searches for allocable idle time slices within the available time window defined by the static gating list, ensuring that the scheduling results of new service flows do not disrupt the timing relationships and forwarding rules determined by the existing static configuration. This achieves incremental scheduling and resource allocation for new service flows while maintaining the stability of the original static gating configuration, thereby supporting dynamic expansion during network operation without affecting the determinism of existing services.
[0050] In a preferred embodiment, incremental gating scheduling calculation includes:
[0051] Based on the network topology information and in accordance with the shortest path principle, the forwarding path corresponding to the new service flow from the source node to the destination node is determined. The forwarding path consists of multiple sequentially connected switching nodes.
[0052] Based on the new business flow, generate the business flow order according to the preset heuristic sorting rules;
[0053] Within the gating list defined by the static configuration data, the idle time slices corresponding to each switching node port through which the new service flow passes during the gating period are extracted. The idle time slices are then traversed according to the sorted service flow order to obtain the scheduling result.
[0054] It should be noted that iterating through the idle time slices includes:
[0055] When the corresponding transmission time can be allocated for the new service flow in the idle time slice of all relevant ports on the forwarding path, an incremental gating scheduling configuration corresponding to the new service flow is generated.
[0056] If any switching node port is unable to allocate an idle time slice that meets the constraints for a new service flow, the incremental scheduling of the new service flow is deemed to have failed.
[0057] As described above, based on network topology information, the forwarding path of new service flows from the source node to the destination node is determined according to the shortest path principle. The network topology information comes from static configuration data pre-issued by the centralized network controller, describing the connection and reachability relationships between switching nodes. By adopting the shortest path principle, the number of forwarding hops can be reduced while ensuring connectivity, thereby reducing end-to-end latency and lowering the complexity of gating scheduling. The resulting forwarding path consists of multiple sequentially connected switching nodes, serving as the basic path constraint for time resource allocation. After determining the forwarding path, for new service flows, a scheduling order is generated according to a preset heuristic sorting rule. The heuristic sorting rule comprehensively considers the resource characteristics of the service flow in the network and its latency sensitivity, such as the path length traversed by the service flow, the data buffer usage of the service flow, the maximum allowed end-to-end latency, and the transmission cycle of the service flow. By sorting the new service flows, services with lower resource usage or higher latency constraints are given priority in scheduling, thereby improving the overall scheduling success rate and reducing the probability of conflicts. After completing the service flow sorting, scheduling... The process enters the time slice allocation phase. Based on the gating list defined in the static configuration data, the switching nodes extract the idle time slices corresponding to each switching node port through which the new service flow passes within a gating cycle. An idle time slice refers to a time window that is not occupied by existing service flows and meets the gating opening conditions. During the scheduling process, the idle time slices are traversed according to the sorted service flow order. When the idle time slices of all relevant ports on the forwarding path can meet the transmission requirements of the new service flow, the corresponding incremental gating scheduling configuration is generated and updated to the relevant switching nodes to achieve timely transmission of the service flow. When any switching node port cannot allocate an idle time slice that meets the constraints for the new service flow, the scheduling of the service flow is determined to have failed, and the failure information is fed back to the sending node or the next-level switching node for scheduling failure handling. Under the premise of meeting the sending cycle, frame length, and timing constraints, an attempt is made to allocate a continuous or corresponding transmission time interval for the new service flow. This realizes the evaluation and scheduling of new service flows one by one without destroying the existing static gating scheduling framework, thereby completing the incremental gating scheduling calculation.
[0058] In a preferred implementation, within the gating period defined by static configuration data, new service flows are sorted and transmission time slots are attempted to be allocated sequentially. The sorting criteria for new service flows include the forwarding path length traversed by the service flow, the data buffer usage of the service flow, the maximum allowed end-to-end latency of the service flow, and the transmission period size of the service flow, thereby allocating idle time slices to the new service flows.
[0059] As described above, when performing incremental gating scheduling, the gating period range and the set of available idle time slices are determined based on static configuration data. On this basis, attribute analysis is performed on the newly added service flows to be scheduled. Attributes include the forwarding path length traversed by the service flow, the data buffer usage of the service flow, the maximum allowed end-to-end latency of the service flow, and the sending period of the service flow. The forwarding path length reflects the number of switching nodes the service flow needs to pass through in the network, used to measure its network resource usage. The data buffer usage reflects the buffer pressure that the service flow may generate in the switching nodes. The maximum allowed end-to-end latency characterizes the real-time requirements of the service flow, and the sending period characterizes the frequency of the service flow's use of time resources. By comprehensively considering multiple dimensions, a ranking relationship between service flows is constructed. In the ranking process, a greedy algorithm is used. The algorithm-based strategy prioritizes new service flows by arranging them one by one, allowing those with lower resource consumption, shorter paths, or higher latency requirements to enter the scheduling sequence first. Idle time slices are then allocated to each new service flow according to this ranking. Specifically, from the idle time slice set defined by the static gating configuration, transmission windows that meet the period and timing constraints of the higher-ranked service flows are prioritized. Once a service flow is successfully allocated a suitable idle time slice, its time resources are removed from the available set. The remaining service flows continue to be scheduled within the updated idle time slice set, thus forming a gradually converging incremental scheduling process. This allows for dynamic access processing without relying on a centralized controller for global replanning, significantly improving the adaptability and scalability of time-sensitive networks in dynamic service access scenarios.
[0060] In a preferred implementation, after completing the incremental gating scheduling calculation, each switching node uploads the scheduling results level by level in a bottom-up manner. When the switching node at the beginning of the forwarding path confirms that the new service flow has been successfully scheduled by all switching nodes on the path, it triggers the collaborative distribution process of the gating scheduling configuration, distributes the updated gating scheduling configuration to the relevant switching nodes via unicast, and confirms that each switching node has completed the configuration update through the configuration response mechanism.
[0061] As described above, after each switching node completes its own incremental gating scheduling calculation, it uploads the calculation result to the next higher-level switching node. This uploading process proceeds from bottom to top along the forwarding path, ensuring that each higher-level switching node can summarize the scheduling status of downstream switching nodes. When the switching node at the beginning of the forwarding path receives the scheduling results from all downstream switching nodes, it determines that the new service flow has been successfully scheduled at every switching node on the path, confirming the effectiveness of the incremental gating scheduling for the entire forwarding path. After confirming the successful scheduling of the new service flow, the starting-end switching node triggers the gating scheduling configuration collaborative distribution process, sending the updated gating configuration to each relevant switching node on the path via unicast. The system configures the data forwarding control based on the latest configuration of each switching node. To ensure the configuration update takes effect, each switching node sends a configuration response signal to the next higher-level switching node after receiving the gating scheduling configuration. This response mechanism confirms that each switching node has completed the configuration update, achieving time-series consistency in configuration effectiveness. It is used for scheduling dynamically added service flows, avoiding the need to re-invoke the centralized network controller, enabling rapid implementation of incremental scheduling, and ensuring the transmission of existing service flows. The system is distributed through bottom-up reporting and unicast, updating the configuration only for the relevant nodes, reducing network-wide broadcasting and centralized controller intervention, and saving communication overhead and computing resources.
[0062] And, a time-sensitive network dynamic gating scheduling terminal, comprising:
[0063] One or more processors;
[0064] A storage device on which one or more programs are stored;
[0065] When one or more programs are executed by one or more processors, the one or more processors implement a time-sensitive network dynamic gating scheduling method.
[0066] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.
Claims
1. A dynamic gating scheduling method for time-sensitive networks, characterized in that, include: The sending node generates new service flow request information and sends the new service flow request information to the directly connected switching node; After receiving a new service flow request, the switching node forwards the new service flow request to the next-hop switching node according to the network topology. When the next hop of the new service flow is the switching node corresponding to the destination node, the node stops forwarding the new service flow request. Any exchange node that receives a new service flow request can perform incremental gating scheduling calculations on the new service flow based on static configuration data pre-issued and stored by the centralized network controller, without re-invoking the centralized network controller. After completing the incremental gating scheduling calculation, the switching node uploads the scheduling result to the next higher-level switching node. When the switching node at the beginning of the path confirms that the new service flow has been successfully scheduled on the corresponding entire forwarding path, it sends the updated gating scheduling configuration to the relevant switching nodes, so that each switching node can perform data forwarding control according to the updated gating scheduling configuration. When any node returns a scheduling failure result, it sends the information that the new service flow cannot be scheduled to the sending node. If the new service flow is successfully scheduled, the switching node directly connected to the sending node sends the corresponding service flow sending time configuration to the sending node, so that the sending node can send data according to the sending time configuration.
2. The time-sensitive network dynamic gating scheduling method according to claim 1, characterized in that, The new service flow request information includes service flow type, source node identifier, destination node identifier, data frame length, number of frames, and sending period parameters.
3. The time-sensitive network dynamic gating scheduling method according to claim 1, characterized in that, Static configuration data is generated and distributed offline by the centralized network controller. During the network initialization phase, the centralized network controller plans the forwarding table entries and gating lists of each switching node in a unified manner based on the global network topology, link resource status, and preset service flow requirements. The generated static gating scheduling configuration is then distributed to each switching node, so that each switching node uses the static gating scheduling configuration as a constraint background when performing incremental gating scheduling calculations.
4. The time-sensitive network dynamic gating scheduling method according to claim 1, characterized in that, Static configuration data includes network topology information, forwarding table information for each switching node port, gating list and periodic information for each switching node port.
5. The time-sensitive network dynamic gating scheduling method according to claim 4, characterized in that, Incremental gating scheduling calculations include: Based on the network topology information and in accordance with the shortest path principle, the forwarding path corresponding to the new service flow from the source node to the destination node is determined. The forwarding path consists of multiple sequentially connected switching nodes. Based on the new business flow, generate the business flow order according to the preset heuristic sorting rules; Within the gating list defined by the static configuration data, the idle time slices corresponding to each switching node port through which the new service flow passes during the gating period are extracted. The idle time slices are then traversed according to the sorted service flow order to obtain the scheduling result.
6. The time-sensitive network dynamic gating scheduling method according to claim 5, characterized in that, Within the gating period defined by static configuration data, new service flows are sorted and transmission time slots are attempted to be allocated sequentially. The sorting criteria for new service flows include the length of the forwarding path traversed by the service flow, the data buffer usage of the service flow, the maximum allowed end-to-end latency of the service flow, and the sending period of the service flow, thereby allocating idle time slices to new service flows.
7. The time-sensitive network dynamic gating scheduling method according to claim 5, characterized in that, Iterate through the idle time slices, including: When the corresponding transmission time can be allocated for the new service flow in the idle time slice of all relevant ports on the forwarding path, an incremental gating scheduling configuration corresponding to the new service flow is generated. If any switching node port is unable to allocate an idle time slice that meets the constraints for a new service flow, the incremental scheduling of the new service flow is deemed to have failed.
8. The time-sensitive network dynamic gating scheduling method according to claim 1, characterized in that, After completing the incremental gating scheduling calculation, each switching node uploads the scheduling results level by level in a bottom-up manner. When the switching node at the beginning of the forwarding path confirms that the new service flow has been successfully scheduled by all switching nodes on the path, it triggers the collaborative distribution process of the gating scheduling configuration. The updated gating scheduling configuration is distributed to the relevant switching nodes via unicast, and the configuration response mechanism confirms that each switching node has completed the configuration update.
9. The time-sensitive network dynamic gating scheduling method according to claim 1, characterized in that, The sending time configuration and gating scheduling configuration adopt a collaborative delivery method for timing control.
10. A time-sensitive network dynamic gating scheduling terminal, characterized in that, include: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors implement the time-sensitive network dynamic gating scheduling method according to any one of claims 1 to 9.